EXPLORE & SUMMARIZE DATA | White Wine Quality Analysis
Aurore Dupont
========================================================
Introduction
In this project, we will be exploring a dataset about the quality of white wines, using exploratory data analysis techniques to explore relationships in one variable to multiple variables in R.
This dataset is public available for research. The details are described in [Cortez et al., 2009].
P. Cortez, A. Cerdeira, F. Almeida, T. Matos and J. Reis.
Modeling wine preferences by data mining from physicochemical properties.
In Decision Support Systems, Elsevier, 47(4):547-553. ISSN: 0167-9236.
Available at:
- Elsevier
- Pre-press (pdf)
- bib
Descriptions of the variables
Input variables (based on physicochemical tests):
- Fixed acidity (tartaric acid - g / dm^3): Most acids involved with wine or fixed or nonvolatile (do not evaporate readily)
- Volatile acidity (acetic acid - g / dm^3): The amount of acetic acid in wine, which at too high of levels can lead to an unpleasant, vinegar taste
- Citric acid (g / dm^3): Found in small quantities, citric acid can add ‘freshness’ and flavor to wines
- Residual sugar (g / dm^3): The amount of sugar remaining after fermentation stops, it’s rare to find wines with less than 1 gram/liter and wines with greater than 45 grams/liter are considered sweet
- Chlorides (sodium chloride - g / dm^3): The amount of salt in the wine
- Free sulfur dioxide (mg / dm^3): The free form of SO2 exists in equilibrium between molecular SO2 (as a dissolved gas) and bisulfite ion; it prevents microbial growth and the oxidation of wine
- Total sulfur dioxide (mg / dm^3): Amount of free and bound forms of S02; in low concentrations, SO2 is mostly undetectable in wine, but at free SO2 concentrations over 50 ppm, SO2 becomes evident in the nose and taste of wine
- Density (g / cm^3): The density of water is close to that of water depending on the percent alcohol and sugar content
- pH: Describes how acidic or basic a wine is on a scale from 0 (very acidic) to 14 (very basic); most wines are between 3-4 on the pH scale
- Sulphates (potassium sulphate - g / dm3): A wine additive which can contribute to sulfur dioxide gas (S02) levels, wich acts as an antimicrobial and antioxidant
- Alcohol (% by volume): The percent alcohol content of the wine
Output variable (based on sensory data):
- Quality: Score between 0 and 10
Several of the attributes may be correlated, thus it makes sense to apply some sort of feature selection.
Univariate Plots Section
We are starting with some preliminary exploration of the dataset.
Summaries of the data and univariate plots will allow us to understand the structure of the individual variables in the dataset.
## X fixed.acidity volatile.acidity citric.acid residual.sugar chlorides
## 1 1 7.0 0.27 0.36 20.7 0.045
## 2 2 6.3 0.30 0.34 1.6 0.049
## 3 3 8.1 0.28 0.40 6.9 0.050
## free.sulfur.dioxide total.sulfur.dioxide density pH sulphates alcohol
## 1 45 170 1.0010 3.00 0.45 8.8
## 2 14 132 0.9940 3.30 0.49 9.5
## 3 30 97 0.9951 3.26 0.44 10.1
## quality
## 1 6
## 2 6
## 3 6- Data have been loaded and the first lines have been displayed to know what the dataset looks like.
## [1] 4898 13- It includes 4,898 observations of 13 variables.
## [1] "X" "fixed.acidity" "volatile.acidity"
## [4] "citric.acid" "residual.sugar" "chlorides"
## [7] "free.sulfur.dioxide" "total.sulfur.dioxide" "density"
## [10] "pH" "sulphates" "alcohol"
## [13] "quality"- Here are the names of each variables (descriptions can be found in the introduction).
## 'data.frame': 4898 obs. of 13 variables:
## $ X : int 1 2 3 4 5 6 7 8 9 10 ...
## $ fixed.acidity : num 7 6.3 8.1 7.2 7.2 8.1 6.2 7 6.3 8.1 ...
## $ volatile.acidity : num 0.27 0.3 0.28 0.23 0.23 0.28 0.32 0.27 0.3 0.22 ...
## $ citric.acid : num 0.36 0.34 0.4 0.32 0.32 0.4 0.16 0.36 0.34 0.43 ...
## $ residual.sugar : num 20.7 1.6 6.9 8.5 8.5 6.9 7 20.7 1.6 1.5 ...
## $ chlorides : num 0.045 0.049 0.05 0.058 0.058 0.05 0.045 0.045 0.049 0.044 ...
## $ free.sulfur.dioxide : num 45 14 30 47 47 30 30 45 14 28 ...
## $ total.sulfur.dioxide: num 170 132 97 186 186 97 136 170 132 129 ...
## $ density : num 1.001 0.994 0.995 0.996 0.996 ...
## $ pH : num 3 3.3 3.26 3.19 3.19 3.26 3.18 3 3.3 3.22 ...
## $ sulphates : num 0.45 0.49 0.44 0.4 0.4 0.44 0.47 0.45 0.49 0.45 ...
## $ alcohol : num 8.8 9.5 10.1 9.9 9.9 10.1 9.6 8.8 9.5 11 ...
## $ quality : int 6 6 6 6 6 6 6 6 6 6 ...- This part shows how the dataframe is structured: either integer or numerical data.
## vars n mean sd median trimmed mad
## X 1 4898 2449.50 1414.08 2449.50 2449.50 1815.44
## fixed.acidity 2 4898 6.85 0.84 6.80 6.82 0.74
## volatile.acidity 3 4898 0.28 0.10 0.26 0.27 0.09
## citric.acid 4 4898 0.33 0.12 0.32 0.33 0.09
## residual.sugar 5 4898 6.39 5.07 5.20 5.80 5.34
## chlorides 6 4898 0.05 0.02 0.04 0.04 0.01
## free.sulfur.dioxide 7 4898 35.31 17.01 34.00 34.36 16.31
## total.sulfur.dioxide 8 4898 138.36 42.50 134.00 136.96 43.00
## density 9 4898 0.99 0.00 0.99 0.99 0.00
## pH 10 4898 3.19 0.15 3.18 3.18 0.15
## sulphates 11 4898 0.49 0.11 0.47 0.48 0.10
## alcohol 12 4898 10.51 1.23 10.40 10.43 1.48
## quality 13 4898 5.88 0.89 6.00 5.85 1.48
## min max range skew kurtosis se
## X 1.00 4898.00 4897.00 0.00 -1.20 20.21
## fixed.acidity 3.80 14.20 10.40 0.65 2.17 0.01
## volatile.acidity 0.08 1.10 1.02 1.58 5.08 0.00
## citric.acid 0.00 1.66 1.66 1.28 6.16 0.00
## residual.sugar 0.60 65.80 65.20 1.08 3.46 0.07
## chlorides 0.01 0.35 0.34 5.02 37.51 0.00
## free.sulfur.dioxide 2.00 289.00 287.00 1.41 11.45 0.24
## total.sulfur.dioxide 9.00 440.00 431.00 0.39 0.57 0.61
## density 0.99 1.04 0.05 0.98 9.78 0.00
## pH 2.72 3.82 1.10 0.46 0.53 0.00
## sulphates 0.22 1.08 0.86 0.98 1.59 0.00
## alcohol 8.00 14.20 6.20 0.49 -0.70 0.02
## quality 3.00 9.00 6.00 0.16 0.21 0.01- We can notice that most of medians and means are close except for residual.sugar.
## wine_id fixed.acidity volatile.acidity citric.acid residual.sugar
## 1 1 7.0 0.27 0.36 20.7
## 2 2 6.3 0.30 0.34 1.6
## 3 3 8.1 0.28 0.40 6.9
## chlorides free.sulfur.dioxide total.sulfur.dioxide density pH
## 1 0.045 45 170 1.0010 3.00
## 2 0.049 14 132 0.9940 3.30
## 3 0.050 30 97 0.9951 3.26
## sulphates alcohol quality
## 1 0.45 8.8 6
## 2 0.49 9.5 6
## 3 0.44 10.1 6- We decided to rename the first column instead having X. New name: wine_id.
## wine_id fixed.acidity volatile.acidity citric.acid residual.sugar
## 1 1 7.0 0.27 0.36 20.70
## 2 2 6.3 0.30 0.34 1.60
## 3 3 8.1 0.28 0.40 6.90
## 4 4 7.2 0.23 0.32 8.50
## 5 5 7.2 0.23 0.32 8.50
## 6 6 8.1 0.28 0.40 6.90
## 7 7 6.2 0.32 0.16 7.00
## 8 8 7.0 0.27 0.36 20.70
## 9 9 6.3 0.30 0.34 1.60
## 10 10 8.1 0.22 0.43 1.50
## 11 11 8.1 0.27 0.41 1.45
## 12 12 8.6 0.23 0.40 4.20
## 13 13 7.9 0.18 0.37 1.20
## 14 14 6.6 0.16 0.40 1.50
## 15 15 8.3 0.42 0.62 19.25
## 16 16 6.6 0.17 0.38 1.50
## 17 17 6.3 0.48 0.04 1.10
## 18 18 6.2 0.66 0.48 1.20
## 19 19 7.4 0.34 0.42 1.10
## 20 20 6.5 0.31 0.14 7.50
## 21 21 6.2 0.66 0.48 1.20
## 22 22 6.4 0.31 0.38 2.90
## 23 23 6.8 0.26 0.42 1.70
## 24 24 7.6 0.67 0.14 1.50
## 25 25 6.6 0.27 0.41 1.30
## chlorides free.sulfur.dioxide total.sulfur.dioxide density pH
## 1 0.045 45 170 1.0010 3.00
## 2 0.049 14 132 0.9940 3.30
## 3 0.050 30 97 0.9951 3.26
## 4 0.058 47 186 0.9956 3.19
## 5 0.058 47 186 0.9956 3.19
## 6 0.050 30 97 0.9951 3.26
## 7 0.045 30 136 0.9949 3.18
## 8 0.045 45 170 1.0010 3.00
## 9 0.049 14 132 0.9940 3.30
## 10 0.044 28 129 0.9938 3.22
## 11 0.033 11 63 0.9908 2.99
## 12 0.035 17 109 0.9947 3.14
## 13 0.040 16 75 0.9920 3.18
## 14 0.044 48 143 0.9912 3.54
## 15 0.040 41 172 1.0002 2.98
## 16 0.032 28 112 0.9914 3.25
## 17 0.046 30 99 0.9928 3.24
## 18 0.029 29 75 0.9892 3.33
## 19 0.033 17 171 0.9917 3.12
## 20 0.044 34 133 0.9955 3.22
## 21 0.029 29 75 0.9892 3.33
## 22 0.038 19 102 0.9912 3.17
## 23 0.049 41 122 0.9930 3.47
## 24 0.074 25 168 0.9937 3.05
## 25 0.052 16 142 0.9951 3.42
## sulphates alcohol quality level
## 1 0.45 8.8 6 Medium
## 2 0.49 9.5 6 Medium
## 3 0.44 10.1 6 Medium
## 4 0.40 9.9 6 Medium
## 5 0.40 9.9 6 Medium
## 6 0.44 10.1 6 Medium
## 7 0.47 9.6 6 Medium
## 8 0.45 8.8 6 Medium
## 9 0.49 9.5 6 Medium
## 10 0.45 11.0 6 Medium
## 11 0.56 12.0 5 Medium
## 12 0.53 9.7 5 Medium
## 13 0.63 10.8 5 Medium
## 14 0.52 12.4 7 Medium
## 15 0.67 9.7 5 Medium
## 16 0.55 11.4 7 Medium
## 17 0.36 9.6 6 Medium
## 18 0.39 12.8 8 High
## 19 0.53 11.3 6 Medium
## 20 0.50 9.5 5 Medium
## 21 0.39 12.8 8 High
## 22 0.35 11.0 7 Medium
## 23 0.48 10.5 8 High
## 24 0.51 9.3 5 Medium
## 25 0.47 10.0 6 Medium- We include a new column indicating if a wine has a Low, Medium, or High quality, such as:
- 1 - 3: Low
- 4 - 7: Medium
- 8 - 10: High
- 1 - 3: Low
Now, let’s plot some univariate data :
- We plotted many histograms/barchart to see if some variables were standing out amongst all of them.
- Log10 transformation have been performed to get a better visualization of the data.
- At first glance, we can say that most of the variables have a fairly balanced distribution (despite some outliers) except for:
- Residual Sugar: It looks like that there are wines with a few sugar in it and other with a lot of sugar.
- Alcohol: The distribution looks unusual.
- This is confirmed by the summary data given above the histograms, when looking at median and mean.
- The normal distribution of each variables makes difficult to determine the ones that have a major impact to predict the quality.
- Most of the wines are rated with a “Medium” quality.
- Quality will be our main feature of interest.
- The boxplots help to check the outliers of each variables.
## wine_id fixed.acidity volatile.acidity citric.acid
## Min. : 1 Min. : 3.800 Min. :0.0800 Min. :0.0000
## 1st Qu.:1225 1st Qu.: 6.300 1st Qu.:0.2100 1st Qu.:0.2700
## Median :2450 Median : 6.800 Median :0.2600 Median :0.3200
## Mean :2450 Mean : 6.855 Mean :0.2782 Mean :0.3342
## 3rd Qu.:3674 3rd Qu.: 7.300 3rd Qu.:0.3200 3rd Qu.:0.3900
## Max. :4898 Max. :14.200 Max. :1.1000 Max. :1.6600
## residual.sugar chlorides free.sulfur.dioxide
## Min. : 0.600 Min. :0.00900 Min. : 2.00
## 1st Qu.: 1.700 1st Qu.:0.03600 1st Qu.: 23.00
## Median : 5.200 Median :0.04300 Median : 34.00
## Mean : 6.391 Mean :0.04577 Mean : 35.31
## 3rd Qu.: 9.900 3rd Qu.:0.05000 3rd Qu.: 46.00
## Max. :65.800 Max. :0.34600 Max. :289.00
## total.sulfur.dioxide density pH sulphates
## Min. : 9.0 Min. :0.9871 Min. :2.720 Min. :0.2200
## 1st Qu.:108.0 1st Qu.:0.9917 1st Qu.:3.090 1st Qu.:0.4100
## Median :134.0 Median :0.9937 Median :3.180 Median :0.4700
## Mean :138.4 Mean :0.9940 Mean :3.188 Mean :0.4898
## 3rd Qu.:167.0 3rd Qu.:0.9961 3rd Qu.:3.280 3rd Qu.:0.5500
## Max. :440.0 Max. :1.0390 Max. :3.820 Max. :1.0800
## alcohol quality level
## Min. : 8.00 Min. :3.000 Low : 20
## 1st Qu.: 9.50 1st Qu.:5.000 Medium:4698
## Median :10.40 Median :6.000 High : 180
## Mean :10.51 Mean :5.878
## 3rd Qu.:11.40 3rd Qu.:6.000
## Max. :14.20 Max. :9.000- As we have seen before, medians and means are quite close, except for residual sugar.
Univariate Analysis
What is the structure of your dataset?
- The dataset consists of 4,898 observations of 14 variables (originally there were 13 variables but I created one more).
What is/are the main feature(s) of interest in your dataset?
- I am interested in looking at the quality of wine, i.e. what chemicals characteristics are most important in predicting the quality.
- We can see that the wines are mostly rated from 3 to 9. Each expert graded the wine quality between 0 (very bad) and 10 (very excellent). However, only a few are very bad or excellent.
What other features in the dataset do you think will help support your
investigation into your feature(s) of interest?
- Acidity like pH, alcohol, density and sugar are characteristics that might help to dig further into the analysis.
- For example, we can ask whether wines with higher alcoholic content receive better ratings, whether sweeter wines contain a higher volume of alcohol or receive better ratings, but also what level of acidity (pH) is associated with the highest quality.
Did you create any new variables from existing variables in the dataset?
- A new variable called “level” has been created for a better readibility of the quality of wine. We can now easily identify if a wine has a bad or high quality.
Of the features you investigated, were there any unusual distributions?
Did you perform any operations on the data to tidy, adjust, or change the form
of the data? If so, why did you do this?
- I created histograms for each variables based on a log10 scale (except for the Quality) to help identify the distribution better.
Bivariate Plots Section
As mentioned in the previous section, there might be some relationships between variables. To look at the correlation, we are creating a matrix.
- The matrix allow us to identify right away what variables are the best to be analyzed.
- We can notice a strong positive correlation between :
- residual sugar & density (0.84),
- free sulfur dioxide & total sulfur dioxide (0.62),
- density & total sulfur dioxide (0.53),
- alcohol & quality (0.44).
- Most of the variables have a negative correlation with quality, which is our main feature of interest.
## Positive Negative
## Small .1 to .3 -0.1 to -0.3
## Medium .3 to .5 -0.3 to -0.5
## Large .5 to 1.0 -0.5 to -1.0- This table help us interpret the correlation coefficient.
- The strongness of the relationships highlighted above are confirmed.
Now, we are analizing the quality versus some variables.
- It seems that the wines with a high quality contain more alcohol.
- However, the density shows an opposite trend. This proves a correlation between alcohol and density.
- Although we can notice a good wine contains less sugar, its trend looks quite flat and doesn’t indicate a major impact on the quality level.
- The pH appears to be a bit greater for wines with a good quality.
- Here, we can see and confirm the relationship between:
- residual sugar & density,
- free sulfur dioxide & total sulfur dioxide,
- density & total sulfur dioxide,
- alcohol & quality.
Let’s take a look at the relationships between the following variables:
- alcohol & pH,
- alcohol & density,
- alcohol & residual sugar.
- We can clearly see that the density decreases based on the volume of alcohol.
- However, this trend is not noticable with the other variables, pH and residual sugar.
- Alcohol vs. Ph scatterplot show that the regression line is quite steady. As for Alcohol vs. Residual sugar one, there a strong decrease in sugar for wines with 8 to 10% of alcohol and then, the volume of sugar is steady as the volume of alcohol goes up.
Bivariate Analysis
Talk about some of the relationships you observed in this part of the
investigation. How did the feature(s) of interest vary with other features in
the dataset?
- Surprisingly, there is not many chemical characteristics that can predict the quality of wine, except for the percentage of alcohol.
- Most of the variables have a negative correlation with quality.
- It appears that a good wine contain a higher volume of alcohol and a higher pH, but the concentration in residual sugar is low as well as the density.
Did you observe any interesting relationships between the other features
(not the main feature(s) of interest)?
- We can notice a strong correlation between residual sugar & density, free sulfur dioxide & total sulfur dioxide, density & total sulfur dioxide, residual alcohol & quality.
What was the strongest relationship you found?
- The strong relationship between % of alcohol and quality cannot be denied.
- However, there is a strong relationship for residual sugar and density with a correlation coefficient of 0.84.
Multivariate Plots Section
- Wines with a better quality and a higher volume of alcohol have less density and less sugar.
- The gap between High and Medium wines is wider for thesugar content.
- The density has a tendency to be lower accross all quality of wines as the volume of alcohol gets higher.
- The pH does not impact a lot the quality of the wine.
- As seen before, wines with a low quality are sweeter and denser.
- Again, the pH doesn’t show a major influence on the quality.
- We can notice that low quality wines tend to be sweeter and denser, which corroborate our previous findings.
- As a reminder, a “good” wine appears to have a higher volume of alcohol.
- We do not identify a strong relationship between pH vs. Alcohol & Sugar.
Now, let’s take a look at our linear model.
##
## Calls:
## m1: lm(formula = quality ~ alcohol, data = wine)
## m2: lm(formula = quality ~ alcohol + sulphates, data = wine)
## m3: lm(formula = quality ~ alcohol + sulphates + pH, data = wine)
## m4: lm(formula = quality ~ alcohol + sulphates + pH + density, data = wine)
## m5: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide,
## data = wine)
## m6: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide +
## total.sulfur.dioxide, data = wine)
## m7: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide +
## total.sulfur.dioxide + chlorides, data = wine)
## m8: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide +
## total.sulfur.dioxide + chlorides + residual.sugar, data = wine)
## m9: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide +
## total.sulfur.dioxide + chlorides + residual.sugar + citric.acid,
## data = wine)
## m10: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide +
## total.sulfur.dioxide + chlorides + residual.sugar + citric.acid +
## volatile.acidity, data = wine)
## m11: lm(formula = quality ~ alcohol + sulphates + pH + density + free.sulfur.dioxide +
## total.sulfur.dioxide + chlorides + residual.sugar + citric.acid +
## volatile.acidity + fixed.acidity, data = wine)
##
## ==================================================================================================================================================================================
## m1 m2 m3 m4 m5 m6 m7 m8 m9 m10 m11
## ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
## (Intercept) 2.582*** 2.341*** 1.683*** -20.991*** -12.729* -23.194*** -21.332*** 111.361*** 122.034*** 109.087*** 150.193***
## (0.098) (0.110) (0.250) (6.181) (6.207) (6.413) (6.419) (13.579) (13.863) (13.507) (18.804)
## alcohol 0.313*** 0.314*** 0.311*** 0.353*** 0.358*** 0.352*** 0.336*** 0.202*** 0.188*** 0.239*** 0.193***
## (0.009) (0.009) (0.009) (0.015) (0.015) (0.015) (0.015) (0.019) (0.020) (0.019) (0.024)
## sulphates 0.476*** 0.429*** 0.392*** 0.360*** 0.425*** 0.432*** 0.670*** 0.659*** 0.576*** 0.631***
## (0.100) (0.101) (0.101) (0.101) (0.101) (0.101) (0.102) (0.102) (0.099) (0.100)
## pH 0.225** 0.229** 0.213** 0.233** 0.215** 0.488*** 0.552*** 0.466*** 0.686***
## (0.077) (0.077) (0.076) (0.076) (0.076) (0.079) (0.081) (0.079) (0.105)
## density 22.367*** 13.858* 24.578*** 23.015*** -110.494*** -121.415*** -108.112*** -150.284***
## (6.093) (6.125) (6.345) (6.347) (13.611) (13.909) (13.552) (19.075)
## free.sulfur.dioxide 0.006*** 0.009*** 0.009*** 0.007*** 0.006*** 0.004*** 0.004***
## (0.001) (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
## total.sulfur.dioxide -0.002*** -0.002*** -0.002*** -0.002*** -0.000 -0.000
## (0.000) (0.000) (0.000) (0.000) (0.000) (0.000)
## chlorides -2.235*** -1.456** -1.594** -0.540 -0.247
## (0.552) (0.550) (0.550) (0.539) (0.547)
## residual.sugar 0.062*** 0.066*** 0.066*** 0.081***
## (0.006) (0.006) (0.006) (0.008)
## citric.acid 0.356*** 0.059 0.022
## (0.096) (0.095) (0.096)
## volatile.acidity -1.896*** -1.863***
## (0.113) (0.114)
## fixed.acidity 0.066**
## (0.021)
## ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
## R-squared 0.190 0.193 0.195 0.197 0.209 0.215 0.218 0.237 0.239 0.280 0.282
## adj. R-squared 0.190 0.193 0.194 0.196 0.209 0.214 0.217 0.236 0.238 0.279 0.280
## sigma 0.797 0.796 0.795 0.794 0.788 0.785 0.784 0.774 0.773 0.752 0.751
## F 1146.395 587.145 394.902 300.301 259.103 223.866 194.831 189.967 170.827 190.448 174.344
## p 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
## Log-likelihood -5839.391 -5828.015 -5823.718 -5816.981 -5779.268 -5760.359 -5752.163 -5691.736 -5684.858 -5548.674 -5543.740
## Deviance 3112.257 3097.833 3092.402 3083.907 3036.780 3013.424 3003.356 2930.157 2921.939 2763.891 2758.329
## AIC 11684.782 11664.029 11657.436 11645.962 11572.535 11536.719 11522.326 11403.472 11391.715 11121.347 11113.480
## BIC 11704.272 11690.016 11689.918 11684.942 11618.011 11588.691 11580.795 11468.438 11463.177 11199.306 11197.936
## N 4898 4898 4898 4898 4898 4898 4898 4898 4898 4898 4898
## ==================================================================================================================================================================================- The R-squared value is 0.282. This means that the variables would have an impact of 28% only on the quality.
Multivariate Analysis
- Alcohol and Density have an impact on the quality of wine.
- The density has a robust relationship with residual sugar.
- We have been able to emphasize that quality increases as alcohol increases and density & residual sugar decrease.
Final Plots and Summary
Plot One | Quality vs. Alcohol
Description One
- Most of the wines are categorized with a medium quality.
- This shows that wines with higher quality tend to have a higher volume of alcohol.
- Alcohol is the variable that has the strongest correlation based on quality.
Plot Two | Quality vs. Alcohol & Quality vs. Density
Description Two
- I want to show the negative correlation between alcohol and density and their impact on the quality.
- Based on the outliers and intervals, density and alcohol do not give better wine if we had to choose one variable only to predict the wine quality.
Plot Three | Residual Sugar vs. Density vs. Quality
Description Three
- This plot summarizes the fact that good wines contain less sugar and are less dense.
Reflection
- We went over a dataset containing 4,898 observations and 13 variables.
- It is surprising that only a few factors (alcohol, density and residual sugar) affect the wine quality.
It is likely that with more variables - such as grape types, states, weather (rain and sun impact the quality), expertise, organic or not, fermentation time, type of barrel,- a thorough analysis could have been conducted.
- Where did I run into difficulties in the analysis?
- At first, it was not clear what variables I had to investigate. That’s why I chose to plot all of them and see if one of them was standing out. Then, the relationships between variables got clearer thanks to the correlation matrix.
- Also, I am not really familiar with chemical compounds, that is why I chose to dig further the following variables Density, pH, Residual Sugar and Alcohol. These terms are more familiar to me.
- Where did I find successes?
- I knew that the percentage of alcohol has an impact on the wine quality. And indeed, it got confirmed through the analysis.
References
https://s3.amazonaws.com/content.udacity-data.com/courses/ud651/diamondsExample_2016-05.html
http://rprogramming.net/rename-columns-in-r/
http://www.cookbook-r.com/Manipulating_data/Adding_and_removing_columns_from_a_data_frame/
https://stackoverflow.com/questions/19440069/ggplot2-facet-wrap-strip-color-based-on-variable-in-data-set
https://cran.r-project.org/web/packages/corrplot/vignettes/corrplot-intro.html
https://stackoverflow.com/questions/21140798/error-using-corrplot/21141799
http://www.sthda.com/english/wiki/correlation-test-between-two-variables-in-r
https://statistics.laerd.com/statistical-guides/pearson-correlation-coefficient-statistical-guide.php
https://www.cyclismo.org/tutorial/R/tables.html
http://r-statistics.co/Linear-Regression.html
https://stackoverflow.com/questions/43359050/error-continuous-value-supplied-to-discrete-scale-in-default-data-set-example/43359104
http://felixfan.github.io/ggplot2-remove-grid-background-margin/
https://rstudio-pubs-static.s3.amazonaws.com/228019_f0c39e05758a4a51b435b19dbd321c23.html#1_plot_one_variable_-_x:_continuous_or_discrete